AES/EBU to S/PDIF converters

[hege]

Let's say I want to convert AES signal to S/PDIF input in some DAC.

Canare has two options:

BCJ-XJ-TRB - passes through level unchanged (5V p-p)
BCJ-XJ-A10TRA - has a 10dB pad (1V p-p)

http://www.canare.com/UploadedDocuments/110Ω-75Ω Digital Audio Impedance Transformers.pdf

Neutrik doesn't have any padding option from what I see.

https://www.neutrik.com/en/neutrik/products/accessories/aes-ebu-impedance-transformers

So which is the most "foolproof" option? Googling seems to suggest that most S/PDIF inputs handle 5V p-p just fine. But the Canare 1V p-p option seems to be more inline with S/PDIF standards? But does it matter in any way, as Neutrik doesn't even bother to offer the option?

 

[Pluto]

I have used both these products with an entirely predictable outcome. You are unlikely to need the Canare padded option which is not for general purpose use, but when you want to interface the coaxial side with a video distribution system*.

Remember that an impedance-changing transformer has the concomitant effect of adjusting the voltage level accordingly. Like Love and Marriage, as the song says, you can't have one without the other.

Of greater significance is the fact that the status bits in pure SPDIF are slightly different to those within AES/EBU. The possible significance of this depends on how strict the implementation happens to be on either side but I doubt you will, in reality, experience any problems. For an illustration of how these status bits were once used, look here.

*If you really need it, you can knock up a suitable attenuator for $1. To undo the effect of unwanted attenuation is considerably more costly.

 

[sergeauckland]

I've only found one product that was sensitive to the Pro/Consumer status bit, it was a broadcast digital mixer than wouldn't accept an S-PDIF stream in its AES-EBU inputs not because of voltages, but purely because of the status bit. Every other piece of digital equipment couldn't care less what was fed into the digital input, whether an S-PDIF source going into an AES-EBU balanced input simply with pins 1&3 strapped, or an AES-EBU output feeding an S-PDIF input by either going between XLR pins 1&2 and strapping pins 1&3. AES-EBU outputs and inputs should use a pulse transformer which allows either pin 2 or 3 to be grounded, but there are some circuits around which use electronic balancing that may not be so tolerant, so it's worth leaving pin 3 floating at first, if it works great, if it doesn't then ground pin 3.

S.

 

[mansr]

Let's say I want to convert AES signal to S/PDIF input in some DAC.

Canare has two options:

BCJ-XJ-TRB - passes through level unchanged (5V p-p)
BCJ-XJ-A10TRA - has a 10dB pad (1V p-p)

http://www.canare.com/UploadedDocuments/110Ω-75Ω Digital Audio Impedance Transformers.pdf

Neutrik doesn't have any padding option from what I see.

https://www.neutrik.com/en/neutrik/products/accessories/aes-ebu-impedance-transformers

So which is the most "foolproof" option? Googling seems to suggest that most S/PDIF inputs handle 5V p-p just fine. But the Canare 1V p-p option seems to be more inline with S/PDIF standards? But does it matter in any way, as Neutrik doesn't even bother to offer the option?

S/PDIF is 0.5 Vpp nominal, 0.6 Vpp max. AES/EBU is 2-7 Vpp nominal. Both are minimum 0.2 Vpp at the receiver.

To be completely within spec, no passive attenuator can work. In practice, you're very unlikely to run into problems whatever you do. You'd have to go out of your way to design an S/PDIF input that actually fails to operate (or breaks) with an input of a few volts, and many ICs are made to handle both types anyway. Impedance matching isn't usually critical either.

 

[Bulldogger]

I used the Canare BCJ-XJ-TRC, the one with AES/EBU to BNC output. I have a Lynx AES-16e card and wanted use my XLR connection on my dac for my Berkeley Alpha USB. It seemed to work fine at first, but my dac started to malfunction when I used it. The volume levels would suddenly ramp up and various controls no longer worked. I plan to use try the Hosa CDL-313. I contacted Canare and was told that was the proper way to do it, with a device like that and the BCJ-XJ-A10TRA was designed for video. I contacted, Hosatech and asked about it being only being specd for up to 24/96 and was told that because that's the highest sample rate for which it had been tested. Higher sampler rates may work. I'll test that shortly.

 

[Catalo]

I’ve been using the Hosa CDL-313 to feed an all-digital chain to the AES input into my JBL 708P speakers for a few months and never noticed any unexpected behavior. PC -> Nuforce UDAC2 doing usb to coax conversion -> Hosa doing coax to AES -> right speaker -> AES through to left speaker. The device converts both ways simultaneously, so if you need AES to coax, you have the option as well.

 

[Pluto]

There are some important distinctions between the Hosa unit and others discussed here insofar as the former is i)active*, ii)bi-directional, iii)claims to make an adjustment to the status bits and iv) can be over twice the price of its passive counterparts.

For the task in hand as stated by the OP

I want to convert AES signal to S/PDIF input in some DAC

KISS!

The simpler, passive option is the way to go.

*and therefore requires a power supply

 

[Bulldogger]

There are some important distinctions between the Hosa unit and others discussed here insofar as the former is i)active*, ii)bi-directional, iii)claims to make an adjustment to the status bits and iv) can be over twice the price of its passive counterparts.

For the task in hand as stated by the OP


KISS!

The simpler, passive option is the way to go.

*and therefore requires a power supply

The passive components do not reduce signal to proper levels.

 

[Pluto]

Professional units such as the Neutrik perform exactly as specified. I have passed hundreds of hours of material through such devices, which do exactly what it says on the tin. Bear in mind that they do not claim to convert to SPDIF specifications; it just so happens that the co-ax output works satisfactorily with most SPDIF inputs.

If the excess level is an issue for you, it's nothing that a 10dB inline BNC attenuator won't solve.

 

[sergeauckland]

Professional units such as the Neutrik perform exactly as specified. I have passed hundreds of hours of material through such devices, which do exactly what it says on the tin. Bear in mind that they do not claim to convert to SPDIF specifications; it just so happens that the co-ax output works satisfactorily with most SPDIF inputs.

If the excess level is an issue for you, it's nothing that a 10dB inline BNC attenuator won't solve.

Bear in mind also that there are two specifications for AES-EBU, the more common (in audio circles) of the 110 ohm balanced, and the 75 ohm unbalanced specification, more common in Video facilities as it can be carried over any old Composite Video cabling which has to be accurately 75 ohms. The differences then between 75 ohm AES-EBU and S-PDIF are in voltage level, (AES-EBU is higher) and the Pro/Consumer status bit which pretty much all AES-EBU and S-PDIF inputs ignore. The audio data bits are identical in AES-EBU and S-PDIF.

S.

 

[Bulldogger]

Professional units such as the Neutrik perform exactly as specified. I have passed hundreds of hours of material through such devices, which do exactly what it says on the tin. Bear in mind that they do not claim to convert to SPDIF specifications; it just so happens that the co-ax output works satisfactorily with most SPDIF inputs.

If the excess level is an issue for you, it's nothing that a 10dB inline BNC attenuator won't solve.

Canare has a 10db attenuator in the model I used. Didn't work. Caused my dac to malfunction. Contacted Neutrik today and told no guarantee it would work either. I'll use the example of my friend's car. His catalytic converter needed replacing. The KISS solution, the mechanic, if you can call him that, used was to take a hole saw and cut a hole in the Cat. That works for some older cars, but not for newer cars. It caused his car to malfunction. The Hosa cdl-313 is a 100% solution and will work with all products because it matches levels and impedance to both standards. It's all I use. 'But hey, you can try the "hole saw" first.